Abstract: The paper presents results of efforts of increasing predicting events accuracy by increasing a set of attributes describing the present moment by information included in past data. There are described two experiments verifying such an approach. The experiments were carried on by the use of the RSES system, which is based on the rough sets theory. The data analyzed in the first experiment, concerning the weather, were reported at the meteorological station in Jasionka near Rzeszów from 1 April 2004 to 30 september 2005. The second experiment deals with exchange rates based on the money.pl news bulletin data (http://www.money.pl).

Keywords: rough sets, prediction, temporal data.

Link:

ROUGH SET METHODS IN ANALYSIS OF CHRONOLOGICALLY ARRANGED DATA

Piotr Romanowski

http://foibg.com/ibs_isc/ibs-23/ibs-23-p08.pdf